Identifying Institutional Diagnostic Reference Levels for CT with Radiation Dose Index Monitoring Software

Purpose: To retrospectively evaluate radiation optimization efforts over 4 years for three computed tomography (CT) protocols and to determine institutional (local) diagnostic reference levels for prospective tracking by using automated radiation dose index monitoring software. Materials and Methods: Approval for this retrospective observational study was obtained from the hospital research ethics board, and the need to obtain informed consent was waived. The study followed a 48-month radiation dose optimization effort in a large academic inner-city trauma and quaternary referral center. Exposure according to equipment, protocol, and year (2010-2013) for adult patients was determined for routine unenhanced head CT examinations, CT pulmonary angiography examinations, and CT examinations for renal colic. Mean exposure (as volume CT dose index [CTDIvol] and dose-length product [DLP]) was averaged to establish local diagnostic reference levels. Means and 75th percentiles for 2013 were compared with findings from surveys in Canada and diagnostic reference levels for similar protocol types internationally. Student t tests were performed to assess significance between annual means, and chi(2) tests were performed for changes in categoric variables. Results: There were 36 996 examinations in 25 234 patients. There was an average exposure reduction of 22% for CTDIvol and 13% for DLP from 2010 to 2013. In 2013, mean CTDIvol for routine head examinations was 50.8 mGy +/- 3.7 (standard deviation), 11.8 mGy +/- 5.6 for CT pulmonary angiography examinations, and 10.2 mGy +/- 4.2 for renal colic CT examinations, while mean DLP was 805.7 mGy . cm +/- 124.3, 432.8 mGy-cm +/- 219.9, and 469.4 mGy . cm +/- 209.2, respectively. The mean CTDIvol and DLP in 2013 were at or close to identified reference values; however, additional optimization is required to reach "as low as reasonably achievable" values for all examinations. Conclusion: Automated methods of radiation dose data collection permit a detailed analysis of radiation dose according to protocol and equipment over time. Radiation dose optimization measures were effective, but their full value may be realized only with changes in internal processes and real-time, prospective data monitoring and analysis. (C) RSNA, 2015